© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 10861
    Prev
    Next
    INDEX
    Explanations

    integration

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron responds strongly to uncommon, content-bearing words (rare nouns or verbs) rather than common function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    тье
    0.81
    真正的
    0.73
    ClNO
    0.72
    γι
    0.72
    倜
    0.70
    DEBUGFS
    0.70
    挑选
    0.68
    ']))->
    0.68
    Ğ
    0.68
     खिलाड़ियों
    0.68
    POSITIVE LOGITS
    بر
    0.72
     Osaka
    0.71
    iritto
    0.71
     fiz
    0.70
     東京都
    0.70
     afferm
    0.70
    ł
    0.70
    ين
    0.69
    كان
    0.69
    عود
    0.69
    Activations Density 0.000%

    No Known Activations